4 ms·

Lambda’s debt-swapping partnership is disappearing from the internet

_bxg1 7y ago
We desperately need an accreditation system for these schools, stat. FAANG continues to scramble for newhires; what if they created a nonprofit that audits and accredits coding bootcamps, to juice up the pool of candidates? Right now it's a massive gamble to sign up for one and hope you're not being scammed. Once perception catches up to reality it might poison the very idea of bootcamps in the public consciousness.
- rahimnathwani 7y ago
  "We desperately need an accreditation system for these schools, stat." I don't know the details, but I like the sound of the approach taken by Make School: https://www.dominican.edu/dominicannews/dominican-announces-computer-science-partnership-with-make-school https://www.dominican.edu/dominicannews/dominican-announces-... In short, Make School partners with an already-accredited university, under a scheme which provides it a path to becoming accredited itself.
- tracerbulletx 7y ago
  We have been hiring Lambda school grads for about a year now and have been really pleased with the success rate. Some really great candidates, many with degrees in the sciences who wanted to transition to CS and chose lambda school. From what I've heard their program sounds pretty good.
throwGuardian 7y ago
I've seen a lot of folks criticizing Lambda. As far as I can tell, they actually offer a chance at a much better life, with no upfront fees - something even accredited and reputable universities are unwilling to offer. If ISAs offer lamda better cash flow that doesn't affect the students in any way, is it really such a bad thing?
- minimaxir 7y ago
  The original article addresses these questions (namely, it's opportunity cost/expected value): https://www.theverge.com/2020/2/11/21131848/lambda-school-coding-bootcamp-isa-tuition-cost-free https://www.theverge.com/2020/2/11/21131848/lambda-school-co...
- _bxg1 7y ago
  This recent investigation (from the same site) is relevant: https://www.theverge.com/2020/2/11/21131848/lambda-school-coding-bootcamp-isa-tuition-cost-free https://www.theverge.com/2020/2/11/21131848/lambda-school-co... > From the beginning, however, the online class wasn’t what Surber or her classmates had expected. The instructors changed week to week and often seemed to have no idea what the students had already covered. The curriculum advertised on the website never fully materialized. The online portal where they were supposed to find their homework assignments rarely matched up with what they were learning. There are good coding bootcamps, and ISAs can work if they're done right, but there's also a whole lot of abuse going on in this space right now.
- ac29 7y ago
  Schools in the US don't have any "upfront fees" either - you can pay for them with student loans that don't begin requiring payments until after you graduate (and in some cases, they don't accrue interest while in school either).
rahimnathwani 7y ago
Assuming investors in these ISAs are rational/smart, they'll take into account the likely outcomes before deciding how much they're willing to pay. So, even if Lambda sells each ISA the moment it's signed, they're still just as incentivised to do a good job. Because the sale price will be determined by the expected cash flows which, in turn, are affected by how good a job Lambda school does. It seems just like any other form of factoring: you've got receivables which are due in the future, and you're willing to take a haircut to get paid sooner. I don't know whether Lambda school provides a good service and/or whether the ISAs they offer are a good deal for the students. But the point of the article is that selling the ISAs or pools of ISAs weakens incentives. I don't see how that follows. If anything, it makes incentives stronger: if Lambda does a bad job, this will affect the money they receive today (via sale of ISAs) rather than money they'll receive over time (from student incomes).
- danso 7y ago
  The WIRED article that Verge references points out a few ways where securitized debt could provide perverse incentives, particularly in terms of cost/quality of the teaching: https://www.wired.com/story/how-we-learn-lambda-income-sharing-agreements/ https://www.wired.com/story/how-we-learn-lambda-income-shari... > Still, that points to a more complicated “incentive alignment” than is often described. Deference to outside investors might encourage a school to issue lots of ISAs and keep instructional costs low, for example, betting high margins on a few successful students will subsidize the other agreements, says Julie Margetta Morgan, a fellow at the Roosevelt Institute who studies student debt. That’s especially true if it proves difficult to link students with jobs. But Allred insists the incentive structure that aligns the interests of individual students with the school’s isn’t broken. “We still hold the bulk of the risk,” he says. Allred doesn’t tell staff which student’s contracts have been financed. That said, I don’t know enough to think that CDOs in themselves are bad, or in any way unusual or suspect. But Lambda and Edly refusing to comment and taking down pages about the CDOs and investments, as DHH tweets about them, makes me very confused and suspicious.
  - rahimnathwani 7y ago
    The example given "betting high margins on a few successful students will subsidize the other agreements" is a possible strategy, but that strategy could be attractive/optimal regardless of whether the ISAs are sold. What about the ISAs being sold to third parties affects the incentives? Whether or not thet sell the ISAa, Lambda school should be trying to maximise the margin between the cashflows from the ISAs, and the cost of running the program.

We desperately need an accreditation system for these schools, stat. FAANG continues to scramble for newhires; what if they created a nonprofit that audits and accredits coding bootcamps, to juice up the pool of candidates? Right now it's a massive gamble to sign up for one and hope you're not being scammed. Once perception catches up to reality it might poison the very idea of bootcamps in the public consciousness.

rahimnathwani 7y ago
"We desperately need an accreditation system for these schools, stat." I don't know the details, but I like the sound of the approach taken by Make School: https://www.dominican.edu/dominicannews/dominican-announces-computer-science-partnership-with-make-school https://www.dominican.edu/dominicannews/dominican-announces-... In short, Make School partners with an already-accredited university, under a scheme which provides it a path to becoming accredited itself.
tracerbulletx 7y ago
We have been hiring Lambda school grads for about a year now and have been really pleased with the success rate. Some really great candidates, many with degrees in the sciences who wanted to transition to CS and chose lambda school. From what I've heard their program sounds pretty good.

throwGuardian 7y ago

I've seen a lot of folks criticizing Lambda. As far as I can tell, they actually offer a chance at a much better life, with no upfront fees - something even accredited and reputable universities are unwilling to offer. If ISAs offer lamda better cash flow that doesn't affect the students in any way, is it really such a bad thing?

minimaxir 7y ago
The original article addresses these questions (namely, it's opportunity cost/expected value): https://www.theverge.com/2020/2/11/21131848/lambda-school-coding-bootcamp-isa-tuition-cost-free https://www.theverge.com/2020/2/11/21131848/lambda-school-co...
_bxg1 7y ago
This recent investigation (from the same site) is relevant: https://www.theverge.com/2020/2/11/21131848/lambda-school-coding-bootcamp-isa-tuition-cost-free https://www.theverge.com/2020/2/11/21131848/lambda-school-co... > From the beginning, however, the online class wasn’t what Surber or her classmates had expected. The instructors changed week to week and often seemed to have no idea what the students had already covered. The curriculum advertised on the website never fully materialized. The online portal where they were supposed to find their homework assignments rarely matched up with what they were learning. There are good coding bootcamps, and ISAs can work if they're done right, but there's also a whole lot of abuse going on in this space right now.
ac29 7y ago
Schools in the US don't have any "upfront fees" either - you can pay for them with student loans that don't begin requiring payments until after you graduate (and in some cases, they don't accrue interest while in school either).

rahimnathwani 7y ago

Assuming investors in these ISAs are rational/smart, they'll take into account the likely outcomes before deciding how much they're willing to pay. So, even if Lambda sells each ISA the moment it's signed, they're still just as incentivised to do a good job. Because the sale price will be determined by the expected cash flows which, in turn, are affected by how good a job Lambda school does. It seems just like any other form of factoring: you've got receivables which are due in the future, and you're willing to take a haircut to get paid sooner. I don't know whether Lambda school provides a good service and/or whether the ISAs they offer are a good deal for the students. But the point of the article is that selling the ISAs or pools of ISAs weakens incentives. I don't see how that follows. If anything, it makes incentives stronger: if Lambda does a bad job, this will affect the money they receive today (via sale of ISAs) rather than money they'll receive over time (from student incomes).

danso 7y ago
The WIRED article that Verge references points out a few ways where securitized debt could provide perverse incentives, particularly in terms of cost/quality of the teaching: https://www.wired.com/story/how-we-learn-lambda-income-sharing-agreements/ https://www.wired.com/story/how-we-learn-lambda-income-shari... > Still, that points to a more complicated “incentive alignment” than is often described. Deference to outside investors might encourage a school to issue lots of ISAs and keep instructional costs low, for example, betting high margins on a few successful students will subsidize the other agreements, says Julie Margetta Morgan, a fellow at the Roosevelt Institute who studies student debt. That’s especially true if it proves difficult to link students with jobs. But Allred insists the incentive structure that aligns the interests of individual students with the school’s isn’t broken. “We still hold the bulk of the risk,” he says. Allred doesn’t tell staff which student’s contracts have been financed. That said, I don’t know enough to think that CDOs in themselves are bad, or in any way unusual or suspect. But Lambda and Edly refusing to comment and taking down pages about the CDOs and investments, as DHH tweets about them, makes me very confused and suspicious.
- rahimnathwani 7y ago
  The example given "betting high margins on a few successful students will subsidize the other agreements" is a possible strategy, but that strategy could be attractive/optimal regardless of whether the ISAs are sold. What about the ISAs being sold to third parties affects the incentives? Whether or not thet sell the ISAa, Lambda school should be trying to maximise the margin between the cashflows from the ISAs, and the cost of running the program.